Introduction
Machine Learning and Artificial Intelligence frameworks are numerous and their impact on the future of computer science need no emphasis. However, deploying these models can be complex and fairly manual without the appropriate multi-function ecosystem, especially when deploying these models to the edge.
Luckily, Cloudera comprehensive data management suite make this endeavor very easy.During this series, I will present how to create a Deep Learning model trained to read digits from the MNIST database and deploy it to the edge.
This article is an introduction to the architecture and pre-requisites necessary for this tutorial. It will refer to sub articles that will be tutorials that anyone can follow to implement understand how to take an AI model and operationalize it to the edge
Architecture overview
The figure below gives a highlight of my hybrid cloud platform:
As you can see, it is comprised of the three main functions:
- Cloudera Data Science Workbench: data science hub used to train and save the model, leveraging like Tensorflow, Jupyter and ONNX.
- Cloudera Flow Management: leverages Nifi to develop a flow reading from an image and running the ONNX model.
- Cloudera Edge Management: allows for the deployment of Minifi flows to edge deployment
Pre-Requisites
To run this tutorial I used the following main elements of the Cloudera Stack:
- CDSW 1.5
- HDF 3.3
- HDP 3.1
- CEM 1.0
Implementation Tutorials
The implementation will be detailed in the upcoming following tutorial articles:
- Part 1: CDSW model training using a custom docker image with Jupyter and save it using ONNX
- Part 2: Nifi flow creation to parse new images and run the model
- Part 3: Flow deployment to Minifi using CEM
